The Role of Tau in Alzheimer's Disease and Related Disorders

SUMMARY

Tau, the microtubule‐associated protein, forms insoluble filaments that accumulate as neurofibrillary tangles in Alzheimer's disease (AD) and related tauopathies. Under physiological conditions, tau regulates the assembly and maintenance of the structural stability of microtubules. In the diseased brain, however, tau becomes abnormally hyperphosphorylated, which ultimately causes the microtubules to disassemble, and the free tau molecules aggregate into paired helical filaments. A large body of evidence suggests that tau hyperphosphorylation results from perturbation of cellular signaling, mainly through imbalance in the activities of different protein kinases and phosphatases. In AD, it appears that ß‐amyloid peptide (Aß) plays a pivotal role in triggering this imbalance. In this review, we summarize our current understanding of the role of tau in AD and other tauopathies, and highlight key issues that need to be addressed to improve the success of developing novel therapies.

Introduction

Alzheimer's disease (AD) is a progressive neurodegenerative disorder associated with memory loss, spatial disorientation, and gradual deterioration of intellectual capacity. Numerous pathological changes have been described in the postmortem brains of AD patients, including synaptic and neuronal loss, oxidative damage, activated inflammatory cells, amyloid plaques mainly composed of the ß‐amyloid peptide (Aß), and neurofibrillary tangles (NFTs) comprised of hyperphosphorylated aggregates of the microtubule‐associated protein tau, the latter two of which are considered the pathological hallmarks [1, 2, 3]. For several reasons, research on the involvement of Aß in AD has progressed more quickly than that on tau. The description of the “amyloid cascade hypothesis” based on the discovery of genetic mutations that cause autosomal familial AD centered the focus of research on Aß[4, 5]. Also, the biochemical studies of amyloid precursor protein (APP) and the presenilins have greatly enhanced the understanding of the molecular pathways leading to Aß generation [6]. These studies favored the systematic development of disease‐modifying therapies based on Aß pathway.

Many clinical trials are currently underway that are evaluating the efficacy of targeting some aspect of Aß biology (i.e., genesis, aggregation, clearance, etc.) in AD patients. Although it is too early to know the outcome of these trials, there is justifiable concern that targeting Aß in patients with AD, even in those with milder stages of the disease, may be insufficient because of the numerous pathways and resultant damage triggered by the accumulation of Aß. Eventually, the field will progress to the point that multiple/combination therapies are evaluated in clinical trials. In the meantime, a challenge for ongoing preclinical studies involving animal models is to identify the most promising therapeutic combinations.

Given its longstanding and prominent role in the molecular pathology of AD, there has been surprisingly scant attention focused on tau as a therapeutic target, which is reflected by the large disparity in clinical trials that target Aß versus tau (see http://www.clinicaltrials.gov). Advances over the past decade in the generating animal models with tau pathology are helping to erase this deficit (Figure 1). In this review, we will consider the existing evidence validating tau as a therapeutic target for AD.

Figure 1.

Key discoveries in tau protein research.

Molecular Biology of Protein Tau

The discovery that mutations in the microtubule‐associated protein tau (MAPT) gene cause fronto‐temporal dementia with Parkinsonism linked to chromosome 17 (FTLD‐17) was a watershed discovery for the field, and provided genetic evidence that established that dysfunction in tau was sufficient to trigger neurodegeneration, in the absence of Aß[7, 8, 9]. Several CNS disorders—AD, Pick's disease (PiD), progressive supranuclear palsy (PSP), corticobasal degeneration (CBD), and argyrophilic grain disease—are characterized by the aberrant accumulation of tau [10]. Hence, developing targets against tau offers the added extrinsic benefit of not only being useful for AD, but also for the larger class of tauopathies.

In the adult human brain, six isoforms of tau are expressed, generated through alternative splicing of the MAPT gene located on chromosome 17q21 [11]. Alternative splicing of exons 2 and 3 yields variants containing zero (0N), one (1N), or two (2N) inserts at the N‐terminus. In addition, the absence or presence of exon 10 leads to tau proteins that contain either three (3R) or four (4R) C‐terminal microtubule‐binding domains, with 3R tau isoforms binding the microtubules less tightly than 4R tau [12]. Hence, six tau isoforms are possible: 3R0N, 3R1N, 3R2N, 4R0N, 4R1N, and 4R2N. In the normal human brain, the expression levels of 3R and 4R tau occur at a 1:1 ratio. Importantly, several tauopathies are characterized by the alteration in the ratio of 3R:4R tau. For instance, some FTDP‐17 families with mutations around exon 10 have an increased proportion of 4R tau, thereby exacerbating the ability of tau to interact with microtubules [7, 8, 9].

The classically described function of tau is as a neuronal microtubule‐associated protein, mainly found in axons. Under physiological conditions, tau exists as a highly soluble and natively unfolded protein that interacts with tubulin and promotes its assembly into microtubules, which helps to stabilize their structure [13]. Recent evidence points to additional functions for tau. For example, tau phosphorylation enables neurons to escape from an acute apoptotic death through stabilizing ß‐catenin [14]. Also, tau exerts an essential role in the balance of microtubule‐dependent axonal transport of organelles and biomolecules by modulating the anterograde transport by kinesin and the dynein‐driven retrograde transport [15, 16]. Mechanistic understanding of the role of tau, including the daunting challenge of elucidating the specific effects of phosphorylation at multiple sites on tau, will be extremely valuable for the future development of treatments for AD and other related disorders.

Tau is regulated during both normal homeostasis and in stress‐induced responses by an array of posttranslational modifications that includes glycosylation, ubiquitination, glycation, nitration, and oxidation. Of the posttranslational modifications, phosphorylation has been most extensively studied (Figure 2). In the healthy brain, 2–3 residues on tau are phosphorylated. In AD and other tauopathies, however, the phosphorylation level of tau is significantly higher, with approximately nine phosphates per molecule [17, 18]. Hyperphosphorylation of tau may occur at different putative serine, threonine, and tyrosine residues through a disruption in the equilibrium of tau kinases and tau phosphatases activity. The consequences of this phenomenon are still under investigation, but produces the outcome of lowering tau's affinity for the microtubules as well as increasing tau's resistance to calcium‐activated neutral proteases and its degradation by the ubiquitin‐proteosome pathway [19]. Ultimately, tau hyperphosphorylation leads to fibrillization and aggregation into NFTs [20, 21, 22, 23, 24]. The major tau kinases include glycogen‐synthase kinase‐3ß (GSK‐3ß), cyclin‐dependent protein kinase 5 (CDK5), cAMP‐dependent protein kinase (PKA), mitogen‐activated protein kinases (MAPK), calcium‐calmodulin‐dependent kinase‐II (CaMK II), and microtubule affinity‐regulating kinase (MARK) [25, 26, 27, 28, 29, 30]. Among the phosphatases, protein phosphatase 2 (PP2A) has been most implicated in the dephosphorylation of abnormal tau [31]. Notably, changes in the expression and/or activation of tau kinases and tau phosphatases have been well documented in AD and related disorders [32, 33, 34, 35]. Studies in transgenic mouse models of AD suggest that it is likely that multiple, overlapping processes contribute to abnormal hyperphosphorylation of tau, including Aß, impaired brain glucose metabolism, and inflammation [36, 37, 38, 39, 40, 41]. Understanding which cellular pathways mediate posttranslational modifications of tau is of high interest in the discovery of therapeutic targets.

Figure 2.

Molecular mechanisms of tau regulation.

Experimental Models of Tau

Since 1995, there has been significant progress in generating cellular and animal models, including using vertebrate and invertebrate species, to analyze and dissect the role of tau in AD and related tauopathies [42]. Because of space limitations, we will focus on transgenic models, and consider how tau transgenic approaches have contributed to the development of novel therapeutic strategies.

There are many studies addressing the pathological effects and functional consequences of human tau mutations, with the most commonly studied tau mutations being glycine residue 272 to valine (G272V), asparagine residue 279 to lysine (N279K), proline residue 301 to leucine (P301L), valine residue 337 to methionine (V337M), and arginine residue 406 to tryptophan (R406W) [43]. All these mutations markedly reduce the ability to promote microtubule assembly, presumably by promoting hyperphosphorylation of tau, followed by its assembly into filaments aggregates. The identification of key kinases and phosphatases that modulate pathological changes of tau in AD and related tauopathies has been another major focus of research. The majority of these studies concentrated on the role that GSK‐3ß and CDK5 play in phosphorylating tau. Double transgenic mouse models have been developed for mutant tau and GSK‐3ß or CDK5 [30, 44]. These bigenic models showed a dramatic increase in tau hyperphosphorylation and NFTs; moreover, the hippocampal atrophy observed in the GSK‐3ß single transgenic mice was accelerated in the tau/GSK‐3ß line [44]. Furthermore, crossing mutant PP2A and tau model resulted in a remarkable increase in the number of the hippocampal neurons expressing hyperphosphorylated tau and NFTs [45]. Together, these data highlight the critical role that certain kinases and phosphatases play in triggering key modifications in tau, and also suggest that these enzymes themselves may be a valuable therapeutic target. However, there are some major concerns about inhibiting kinases and phosphatases because of the plethora of targets they affect. Hence, convincing preclinical data will be required, including in higher order species like nonhuman primates, to ensure that adverse effects are minimized in human trials.

Because APP and/or presenilin mutant transgenic mice do not develop NFT pathology despite extensive plaque deposition, our laboratory adopted an aggressive genetic approach to produce a transgenic model with both plaques and tangles. These mice express the P301L mutation in tau as well as mutant APP and PS1 and are referred as triple transgenic model (3xTg‐AD) [46]. The 3xTg‐AD mice represent a critical resource for elucidating the relationship among Aß, tau, and synaptic dysfunction; they have also proven to be useful for evaluating the efficacy of novel therapeutic strategies in mitigating the neurodegenerative effects mediated by both AD signature lesions. These mice progressively develop Aß and tau pathology, with a temporal‐ and regional‐specific profile that closely mimics their development in the human AD brain [41]. Also, despite equivalent expression of the human APP and human tau transgenes, Aß deposition develops prior to the tangle pathology. Notably, the clearance of Aß using Aß‐specific antibodies reduces the tau burden in the 3xTg‐AD [41]. This is in accordance with studies from others that have shown that intracranial Aß injection in tau mutant mice [47] or coexpression of mutant APP and tau leads to tau pathology [48]. Such studies clearly indicate a link between Aß and tau pathology and suggest that Aß accumulation can exacerbate tau pathology in the course of AD.

The evidence that Aß induces tau phosphorylation has lead to the question about whether the modulation of tau would also influence Aß pathology. Recent study has shown that reduction of tau expression prevents the cognitive deficits in the mutant APP without changing the level of Aß[49]. Moreover, we have shown that genetically augmenting tau levels and hyperphosphorylation in the 3xTg‐AD mice has no effect on the onset and progression of Aß pathology [50]. Together, these studies suggest that the link between Aß and tau is predominantly if not exclusively unidirectional, which is consistent with the Aß cascade hypothesis and may explain why tauopathy‐only disorders are devoid of any Aß pathology.

Further important issues that were recently highlighted using different in vitro and in vivo models regard about how the different tau complexes (soluble oligomeric forms of phosphorylated tau vs. insoluble NFTs) and the extracellular tau may exert their toxicity in the brain. SantaCruz and colleagues [51] have shown that turning off expression of the mutant tau transgene, in a repressible tau transgenic mouse model, attenuates the neurodegeneration and improves memory, in spite of ongoing NFTs accumulation. Supporting this finding, it has been shown that inhibition of tau hyperphosphorylation in the tau P301L transgenic mouse model prevents severe motor deficits and reduces the amount of soluble tau aggregates without changes of NFTs [52]. Also, it has been shown that phosphorylation of soluble tau inhibits the assembly of microtubules by a process dependent on normal MAPs sequestration and results in dendrites and axonal degeneration [53, 54]. These studies strongly suggest that abnormal hyperphosphorylation of soluble forms of tau plays a critical role in the development of tauopathies and that accumulation of NFTs occur independent of memory loss as a possible neural protective mechanism. Concerning the accumulation of extracellular tau, it has been demonstrated by in vitro studies that tau aggregates can propagate a fibrillar, misfolded state from the outside to the inside of a cell [55]. The in vivo relevance of these data is still unknown, however, could represent a new component in the progression of tau dysfunction. Indeed, additional studies are necessary to access the mechanisms in which the different tau filaments and/or oligomeric forms as well as the extracellular tau aggregates affect the neuronal function and exert their toxicity. Strategies that neutralize the intracellular soluble tau forms and the extracellular tau aggregates may be an alternative therapeutic approach for AD and related tauopathies.

Numerous transgenic animal models of AD have been used as important tools for the characterization of abnormal APP processing and/or tau phosphorylation, as well as for the study of the mechanisms of cognitive impairment [42]. Although there is no doubt that these models have helped in unraveling disease processes and have boosted drug discovery and development, they may present an incomplete perspective of AD pathology since most of then employ mutant tau. At present, only a few studies have been developed using transgenic mouse models that express human wild‐type tau protein. Although these mice do not manifest NFTs, they reproduce some features of human pathology such as a strong somatodendritic hyperphosphorylated tau [56, 57, 58, 59, 60, 61]. In addition, Zilka and colleagues have shown that nonmutated truncated tau is an important upstream factor that induces neurofibrillary neurodegeneration of AD type in absence of tau mutation [62]. Future studies will be needed to identify conditions that are able to produce robust pathological states of tau pathology in human wild‐type tau transgenic models.

Tau As a Therapeutic Target

A prominent number of studies focusing on tau therapeutics are in progress. The general information about the principal therapeutics compounds used in the different strategies for tau is summarized in Table 1. Thus, the principal strategies targeting tau in neurodegenerative diseases are (1) to inhibit the abnormal tau hyperphosphorylation through modulation of specific protein kinases as GSK‐3ß, CDK5, casein kinase‐1 (CK‐1), PKA, CaMK II, and MAPK family (ERK1/2, p70S6 kinase, JNK, p38), (2) to induce disassembly of tau aggregates with compounds such as Methylene blue, Antharaquinones, etc., (3) to stimulate microtubule stabilizing molecule with Taxol and Taxol‐derived compounds (taxotere, paclitaxel), (4) to trigger intracellular clearance pathways such as the ubiquitin‐proteasome and/or autophagic system, (5) tau immunotherapy, and (6) antiinflammatory therapy.

Table 1.

Therapeutical tau progress

Antiphosphorylation approaches
Cdk5 inhibitory peptide	Interferes with cdk5/p25 complex formation and inhibits abnormal tau phosphorylation in cortical neurons.	[92, 93]
Roscovitine	A cdk5 inhibitor that reduces tau phosphorylation and neurodegeneration in p25 transgenic mice.	[94, 95]
Lithium	A GSK3β inhibitor that prevents tau phosphorylation, aggregation and axonal degeneration in transgenic mice. Also, in culture cells lithium reduces tau phosphorylation and promotes microtubule assembly.	[96, 97, 98, 99, 100, 101]
AR‐A014418	A GSK3β inhibitor that decreases insoluble and aggregate tau levels.	[102]
AF267B	A selective muscarinic M1 receptor agonist that decreases the level of tau phosphorylation through reduction of GSK3β activity.	[36]
A‐582941	A selective α7‐nAChR agonist that stimulates phosphorylation of GSK3β in residue Ser‐9 and decreases tau phosphorylation in GSK3β‐sensitive and double tau‐APP model.	[103]
NP‐12	A GSK3β inhibitor that reduces tau phosphorylation, neuronal loss in hippocampus and entorhinal cortex, improves spatial memory deficit, and reduces amyloid plaque load in mouse brain.	[104]
K252a	A serine/threonine protein kinase inhibitor that blocks PHF‐like tau hyperphosphorylation in culture of cells and rat brain slices.	[105]
SRN‐033–556 (K252a analog)	Prevents motor deficits and reduces soluble aggregated hyperphosphorylated tau in P301L and JNPL3 transgenic mouse model.	[52]
Memantine	Protects culture neurons against Aβ‐induced toxicity by attenuating tau phosphorylation and restores PP‐2A activity. Chronic memantine treatment reduces the accumulation and tau phosphorylation in the 3xTg‐AD mice. Also, P‐tau is reduced in CSF of AD patient after memantine treatment.	[63, 64, 106, 107]
Docosahexaenoic acid	Reduces the levels of early‐stage phospho‐tau epitopes through inhibition of c‐Jun N‐terminal kinase (JNK).	[69]
Antiaggregation studies
Methylene Blue	A noneuroleptic phenothiazine that reverses the proteolotic stability of tau aggregation and prevents the further propagation of tau capture in AD.	[70]
Anthraquinone	Inhibits tau aggregation and dissolves paired helical filaments (PHF) in vitro and culture assay.	[108]
Phenylthiazolyl‐Hydrazide	Inhibits tau aggregation and disassembles preformed aggregates.	[109]
Rhodanine	Inhibits tau aggregation and promotes paired helical filament (PHF) disassembly in cell model of tauopathy.	[110]
Microtubule‐stabilizing drugs
Taxol	Stimulates microtubule polymerization, stabilizes microtubules, and protects axonal microtubules from accumulating filaments.	[111]
Taxotere	Stimulates microtubule assembly and stabilization. Reverses the effect of tau on mitochondrial distribution.	[112]
Paclitaxel	Stabilizes microtubules and reverses fast axonal transport deficits in a tauopathy model.	[113]
NAP (AL‐108)	A neuronal tubulin‐preferring agent that reduces the levels of soluble and insoluble P‐tau and enhances cognitive function in 3xTg‐AD mice.	[72]
Tau degradation mechanisms
Rapamycin	Stimulates autophagy leading to decreased tau toxicity and decreases insoluble tau in a drosophila fruit model.	[114]
Puromycin‐sensitive aminopeptidase	Mediates amino tau‐degradation of soluble and insoluble (PHF) tau purified from AD brain in vitro studies.	[115]
Tau‐immunotherapy
Active immunization	Active immunization with a phosphorylated tau epitope, in P301L tangle model mice, reduces aggregated tau in the brain, and slows progression of the tangle‐related behavioral phenotype.	[116, 117, 118]
Antiinflammatory approaches
PMX205	Acyclic hexapeptide C5a receptor antagonist reduces hyperphosphorylate tau in 3xTg‐AD mice.	[37]
Atorvastatin	A statin with antiinflammatory properties that reduced the level of neurofibrillary tangle (NFT) burden in DM‐Tau‐Tg.	[67]
Antioxidative studies
α‐tocopherol	Antioxidative agent that suppresses and/or delays the number of tau aggregate in T44‐Tg.	[68]

One of the prominent tau strategies under intensive investigation is to attenuate the abnormal hyperphosphorylation of tau through inhibition of the different tau kinases or restoring or upregulating phosphatase activity, such as PP2A. Along these lines, a recent study showed that memantine (Namenda) restores PP2A activity and reduces abnormal hyperphosphorylation of tau [63, 64]. In addition, we have recently demonstrated that chronic memantine treatment markedly increased the inhibitory phosphorylation of GSK‐3ß and consequently reduce the accumulation and phosphorylation of tau in the 3xTgAD mice [65]. Finally, there is evidence that memantine reduces phospho‐tau species in the cerebrospinal fluid (CSF) of AD patients [63, 64, 66]. The therapeutical use of memantine has been approved only for patients with moderate‐to‐severe AD, as studies in mild‐to‐moderate AD have not consistently revealed significant benefits in this patient population. Further clinical studies are necessary to confirm the therapeutic potential of memantine as a disease‐modifying drug for AD.

A search of the National Institutes of Health (NIH) clinical trial database indicates that over 500 clinical trials are or have been conducted studying some aspect of different tauopathies (search for “tauopathy” in the http://www.clinicaltrials.gov). The majority of these studies, however, are primarily directed to find new AD treatments or diagnostics. A few new classes of drugs that have been previously tested in preclinical tau models are currently being tested in different clinical trial phases, including atorvastatin (NCT00151502) [67], α‐tocopherol (NCT00235716) [68], and docosahexaenoic acid (NCT00440050) [69]. Interesting, the only drugs that are being currently evaluated in clinical trials and that directly modulate tau are Methylene blue (Rember) [70] and the neuronal tubulin‐preferring agent (AL‐108) (NCT00505765) [71, 72]. Methylene blue (Rember) is a compound that belongs to the phenothiazine family, and it is believed that its mechanisms of action occurs by preventing tau aggregation, thereby blocking tau filament formation without an effect on the ability of tau to interact with microtubules. A phase II study has been completed with Rember and a significant improvement in the cognitive function was obtaining compared with placebo patients [73, 74]. A phase III study is planned to validate the efficacy and safety of this compound. The second component is a neuronal tubulin‐preferring agent NAP‐(AL‐108) that promotes microtubule assembly, reduces soluble and insoluble tau hyperphosphorylation, Aβ accumulation, and improved in cognitive memory in 3xTg‐AD [71, 72]. NAP is currently in a human AD phase II study [74, 75]. These studies emphasize the idea of multiple pathways control the progression of the disease and suggest that effective therapeutic approaches will depend on a synergistic combination of different agents to reduce the progression the global impact of these neurodegenerative diseases.

Conclusion Remarks

Over the past few years, evidence has accumulated that point to tau as being a feasible therapeutic target for AD and related disorders. Despite the setbacks associated with developing an effective AD treatment so far, there is reason to be optimistic, as we expand the potential targets to include other key elements of the disease such as tau pathology. The data summarized here indicate that there are several agents that are capable of effectively blocking tau phosphorylation or dysfunction in animal and cellular models. Our predication is that the most likely therapeutic regimen for AD will depend on the use of polypharmacy, that is, a combination of drugs that interfere with different targets in AD, such a blocking the effects of Aß or the subsequent oxidative and inflammatory damage. The successful combination will be assembled stepwise and driven by efficacy testing in one or more of the animal models described here as well as in new models of nonmutated APP and tau. Such studies could not only further our knowledge of tau in AD and related diseases, but also open new avenues for therapy of human dementia.

Conflict of Interest

The authors state that they have no conflict of interest pertaining to this manuscript.